Discussion on 'Two-Stage Procedures for High-Dimensional Data' by Makoto Aoshima and Kazuyoshi Yata

摘要

Professors Mokoto Aoshima and Kaziyoshi Yata have considered several types of statistical inference problems for multivariate high-dimensional data in the so-called large p, small n situation. For this purpose, in their article they have developed the theory for asymptotic normality when p→∞ Their work opens up new directions for further investigations and generalizations. We briefly comment on these and discuss a multinomial problem.